I recently quit a triple-A development house out of frustration; but, instead of going indie, I found a smaller developer that does a broader array of projects. So, instead of working on one FPS for three years, they do several facebook/social/casual games a year.
While that sounds like "more of the same problem", it's actually turned out in my favor. I just managed to get one of my projects greenlit by the executive team. Meaning, we're building something I'm totally in love with!
Now, that's not really answering your question; but, it could prove another avenue for thought. Finding a good developer to work for is just like trying to discover your life long partner. You have to find a company you love... or found one.
As for going the indie route:
- Plan for it -- you're going to need to have money to live off while you're not working; or, you're going to have to do freelance work. Build those funds and relationships now. Set a date and execute it. Possibly even tell your company that you're unhappy and looking to go indie, maybe you can pitch working two or three days a week to them.
- Get yourself up and running -- I know Jari said to get a team; but, honestly? If you can go it yourself, you're far more likely to finish the project. Pull in as few people as you can and, preferably, find individuals that are also fully invested (not working at a large developer most of the time).
- Plan your project -- don't just jump ship and say, "Great! Now I can build something!" You've done Android development and iOS development. Awesome. You know what you're going to have to code. What's your spec? What's your deadline? What product do you want to build? What platforms are you deploying to? Have a plan for success. Do you want to pitch a demo of your game to a publisher? Do you want to compete in an indie competition?
You have a lot of soul searching to do; but, the end result could be spectacular. The way I always looked at it was this: the worst case is that you've just boosted your resume.
Good luck!